Hundreds of empirical studies exploring the relationship of religion to health have been published since the late 19th century. For traditional investigators in biomedicine, this research seems soft since it rarely conforms to the "gold standard" of biomedical research — randomized clinical trials. Even researchers like Jeffrey Levin, Ph.D., a passionate advocate of research on religion, aging, and health, have called some of the empirical work mediocre, even dreadful.
Even so, some scientists concede that there are demonstrated relationships between some religious behaviors and some health outcomes. These links are evident in the areas of cardiovascular disease, hypertension and stroke, cancer (especially uterine cancer), among others. According to Levin, previous studies are remarkably consistent in two areas: (1) There is greater health and less morbidity among such "behaviorally strict" groups as Seventh-day Adventists and Mormons; and (2) the greater the intensity or degree of religiousness (worship attendance and the like), "the better the health and the less of whatever illness is being investigated."
The emerging agreement that there is some connection between religion and health does not, however, eliminate the difficulties in studying this relationship. There are three problem areas. The first is determining what counts as a religious or spiritual intervention: What aspects of religion and spirituality (such as attendance at religious services, following religious dietary habits, engaging in spiritual practices such as meditation) will be studied? How broadly should one define religion and spirituality? How can one quantify religious or spiritual behavior? How does one measure religious commitment?
The second area is determining what counts as a good health outcome. Specifically, must an outcome result in changes in a biomedical category (such as blood pressure or wound healing) to count? Must relaxation exercises for borderline hypertensive patients contribute to lowered blood pressure, or is it sufficient when the person reports a generalized sense of peace and contentment? The benefits that people experience from spiritual practices might not "count" for much in biomedical terms but they may mean a great deal to the patient.
The third area is about method. Designing good studies that prove cause-effect relationships between an intervention and an outcome is tricky even for traditional biomedical interventions. Imagine the difficulties for non-traditional interventions. Let us assume that investigators have solved the problems of how to measure specific religious or spiritual practices, behaviors, or interventions, and have also determined what outcomes they are measuring. How then do they determine if the particular practice has the effect or outcome sought? How does one handle the problem of control groups? For example, while a researcher can measure the effects of intercessory prayer done by a third party on the recovery rates of patients who had cardiac bypass surgery, it is almost impossible to know if other patients—the control group—also had prayers said for them.
Even if randomized clinical trials can be adapted to meet the needs of non-traditional interventions, some practitioners of alternative and complementary medicine strongly oppose such trials on philosophical grounds. For them, alternative therapies represent a different way of understanding the body and its relationship to mind and spirit than more traditional allopathic medicine. This different understanding requires another way of exploring the relationship between cause and effect. It is hard to know what shape these designs will take. Will they involve qualitative studies with interviews or participant-observer research, or will they entail totally new study designs?
American interest in alternative and complementary medicine, especially the relationship of religion and spirituality to health, means that the discussion will continue to widen over the next few years. If this discussion is to move forward, it must face and address value questions. If research into alternative and complementary medicine is to affect biomedicine (and the funding that accompanies it), it must either fit into that mold or develop research strategies that are different but equally compelling. In either case, value questions cannot be avoided.
Martha Holstein is a researcher at the Park Ridge Center.
